论文信息 - A Framework of GRID Problem-Solving Environment Employing Robust Evolutionary Search

A Framework of GRID Problem-Solving Environment Employing Robust Evolutionary Search

This paper presents a problem-solving framework based on robust evolutionary search in GRID computing environment. Our problem-solving environment called virtual innovative laboratory performs simulator programs in parallel and optimize their input parameters employing a competent evolutionary algorithm with gene analysis. The objective of our project is to replace a part of human designer's try-and-error processes by a parallel and robust evolutionary search on GRID computing systems.

[1] Masaharu Munetomo,et al. A Parallel Genetic Algorithm Based on Linkage Identification , 2003, GECCO.

[2] William Gropp,et al. Optimization environments and the NEOS server , 1997 .

[3] Ian T. Foster. Globus Toolkit Version 4: Software for Service-Oriented Systems , 2005, NPC.

[4] 辻美和子. Designing genetic algorithm based on exploration and exploitation of gene linkage , 2007 .

[5] Ian Foster,et al. The Grid 2 - Blueprint for a New Computing Infrastructure, Second Edition , 1998, The Grid 2, 2nd Edition.

[6] Elizabeth D. Dolan,et al. NEOS Server 4.0 Administrative Guide , 2001, ArXiv.

[7] Jorge J. Moré,et al. The NEOS Server , 1998 .

[8] Masaharu Munetomo,et al. Identifying Linkage Groups by Nonlinearity/Non-monotonicity Detection , 1999 .

[9] Thomas Stützle,et al. SATLIB: An Online Resource for Research on SAT , 2000 .

[10] Julian F. Miller,et al. Genetic and Evolutionary Computation — GECCO 2003 , 2003, Lecture Notes in Computer Science.